- Jul 19, 2018
- Midwest, USA
It's a real thing, but isn't caused by any one specific driver behavior. It is a catch-all term to sum up several different driver behaviors that contribute to the perception of "speed" or lack thereof. This is true for many of these "audiophile" terms and "speed" is probably one of the less egregious ones. I personally hate the word "timbre" or "good timbre" because it literally encompasses "everything about the headphone's sound that's not tonality" and says nothing about its performance or even any specific description of sound.
For wired cans, I would far rather them build and ship them as close as they can to the sound they want and suggest EQ or build plugins like audeze. I wouldnt want a world where bluetooth hp became synonymous with "better".
First time to see headphone's sigle point THD+N graph, very interesting. can this be a reasonable explaination of the unique bass sound impression provided by HD6xx series?
I kinda feel like you do, I've been using them practically the entire time I've had them with a custom, unit specific EQ from Sonarworks, and I've absolutely loved them. For me they have been absolutely great for the kind of 60's and 70's rock and 50's jazz I love.Would appear so- I am not sure at all this is a good thing, but as long as readers understand that is one of the key deciders for the panther- so be it.
What I find interesting is that from a technical perspective (distortion etc) its great- but from a stock tuning angle- dud.
Which makes me wonder @amirm, should there be a two tier rating system? Most here know that EQ is pretty essential to get the best out of almost all transducers. A headless panther feels right for the stock performance, but not for what this (or perhaps any) HP can do.
Maybe headless should be reserved for a HP that is neither well made, bad FR as stock and also cant easily be EQ'd to something far closer to target without distorting?
Im overly complicating and as a subjective fan of the LCD-X should just take it on the chin .
May I ask you to elaborate a bit on what is modified smpte IMD 40 & 7000HZ 4:1 ratio? Again out of personal interest, I get you want to asses how loud it can go, but I'd like to know a bit more about what we are looking at. Don't worry we'll get it, It's the first full measurments of headphone here I think, so it's great to have a bit of insight on what you are measuring. In all honesty, I don't even know what smpte ratio mean so if you could quickly walk me trough that?
The standard test tones are 60 and 70 kHz. 60 Hz lands on mains frequency in US so I choose to move that to 40 Hz and hence the "modified" part. The 4:1 refers to the amplitude of the two tones. 40 Hz is four times stronger than 7 kHz.May I ask you to elaborate a bit on what is modified smpte IMD 40 & 7000HZ 4:1 ratio?
It is "THD+N" but for two tones.In all honesty, I don't even know what smpte ratio mean so if you could quickly walk me trough that?
I suspect the hp reviews will evolve but I must admit I think overall efforts are best expended in other areas. But it's a big area of interest and growth so maybe its worth Amir having a chunk of the pie.I strongly agree. Unlike the other ASR content, from which I learn a lot, I learned literally nothing about the LCD-X that I didn't already know from other sources, and in particular the single frequency distortion measurements at only a single SPL aren't very useful.
Can you include these full frequency sweeps by default in future reviews?I would caution against putting any trust in distortion measurements from Tyll. This wasn't his forte and there is no evidence that his measurements are correct. Here are the THD levels I have at the same level that the frequency response was measured for LCD-X (different than the dashboard):
View attachment 88133
As you see, there is no dip at all at 40 Hz. It wouldn't make sense to have a dip in distortion anyway. But rather, you are falling victim to misleading way THD percentage works in systems with varying frequency response like speakers and headphones. There, the amplitude changes and that causes THD% to vary even though the distortion mechanism itself does not. Here is Tyll's frequency response test that shows this:
View attachment 88134
Notice as I circled, his measurements show sudden increase in response around 40 Hz. So for the same distortion amount, the percentage will shrink.
In my testing I am using a single tone @40 Hz and assuring that the identical reproduction level is used in both headphones. As a result, frequency response variations are dialed out and THD ratios can be compared.
Is there a better way to visualize harmonics than a single frequency?
Thanks. This is helpful.Distortion levels are frequency and above all amplitude (level) dependent.
One also needs to take audibility levels and actual listening levels using music into account.
The difficulty with measuring is background noise. That's why on Tyll's plots 90 dB is worse than 100dB where in reality 90dB will be better but only appears to be worse.
To get down really low you need a completely dead room, both in reflections as well as background noises which for the lowest frequencies is extremely difficult. Otherwise it becomes just 'an indicator' for gross errors (which in my book is fine).
So it helps if distortion levels are measured at different SPL (realistic ones) and over the entire band (depends on the harmonics as well)
This means perhaps FFT at various frequencies and amplitudes (that means a LOT of measurements and heaps of plots and posting them).
Too bad if you have one at 5kHz for instance where it happens to be really bad but excellent at 4.5 and 5.5kHz for instance and almost inaudible with music.
The 40Hz at 100dB SPL will end really really bad for most dynamic headphones yet may sound totally fine at 90dB average SPL levels (is really loud already) with music peaking at 100dB SPL.
The review from Amir is a quick one with FR (as it happens to be for that particular rig only) acc. to a standard where the SINAD number will be no measure for quality with a different distortion measurement and Amir's short subjective opinion.
Based on that property alone the LCD-X would get the highest recommendation and be top of the list warranting another panther.
The FR plot is really only readable for most initiated people who can 'translate' the curve + target in their heads to estimate FR. so would expect the 'corrected' version as well for the folks that have a hard time correlating (includes me) to actual sound.
With EQ you can 'fix' a lot of headphones and improve others. It is essentially needed for almost all headphones.
It doesn't look like the HP measurements will be a 'one stop shop' type of thing just like it isn't on all other websites. These too only have scattered info someone might be looking for. Rtings also adds info, crinacle adds info, oratory adds info even RAA adds other info.
I see Amir's review as an addition of the info that is already available. A second opinion, another few reference points to a giant puzzle as it were.
Just made with other equipment and another copy of an already measured headphone elsewhere.
It can add info or confirmation of what's already known. If that's what Amir is after then that's O.K.
At this moment it is nowhere near the Klippel measurements and far removed from the DAC and amp measurements.
Amir is just starting this so give him some slack and pointers to fine tune his thing and am interested to see how this evolves and what the standard 'reporting' will be and how this will add info to what's already out there.
Just measuring and reporting FR on one particular test rig isn't all the info that is needed. The most important factor... yes. But not the only one.
The need to add EQ may be insignificant in some scenarios, and it may be a problem in others. With Amazon HD specifically, and analog, it may be a problem you have to solve, not just a simple implementation.Weird how quick audiophiles will bin stuff that don't match their views. Since one bit I was at few were fuming at that the 5% of the HD650 & how it fine, One raging had a GS3000e?. Won't change that 0.15 ~ 1% for bass going to sound much cleaner than 5 ~ 20% that affects the overall sound, Which can be tested with EDM or anything rich under 100Hz. HD650 high THD at 40Hz seems like the driver struggling to do bass with mids/treble, Which could be from diaphragm stress or it being a old 40mm driver.
The ER3/ER4 uses BA that can be <0.35% under 200Hz on many charts, The Grado/HD650 fanboyism is getting annoying. Since if 1% at 1KHz a issue then i can say the HD650 being 5% and any Grado being 15% is one too.
The need to add EQ may be insignificant in some scenarios, and it may be a problem in others. With Amazon HD specifically, and analog, it may be a problem you have to solve, not just a simple implementation.
I agree with you on HD600 series. Nothing special to my ears, and distortion is not the biggest flaw. They sound their best with my Darkvoice 336SE, which improves FR and adds distortion.